Model profile
GPT Oss 20B
Evidence summary
GPT Oss 20B has an estimated overall rank of #107; its 90% source-sensitivity interval is #52–#223. Its behavior-only rank is #117; company governance moves the combined estimate to #107. Published evidence spans 16 evals and 7 of 7 behavior components. Its strongest relative result is HELM Safety (simple_safety_tests, #1 of 80); its weakest is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #220 of 239).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗gpt-oss-20b
- Hugging Face ↗openai/gpt-oss-20b
- NVIDIA ↗qc69jvmznzxy/gpt-oss-20b
- OpenRouter ↗openai/gpt-oss-20b
- Model card ↗Family-level model document · OpenAI · first party
- Release source ↗direct research preview identity date
Published eval results
Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #217 / 330 | ↓0.8671 | Source ↗official | |
| AIRBench 2024 Safety Scenariossafety_scenarios | #10 / 80 | ↑0.86 | Source ↗official | |
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #30 / 104 | ↓14.17 | Source ↗official | |
| DystopiaBenchbasaglia_score | #42 / 50 | ↓70.13 | Source ↗official | |
| DystopiaBenchbaudrillard_score | #37 / 50 | ↓66.3 | Source ↗official | |
| DystopiaBenchhuxley_score | #37 / 50 | ↓75.47 | Source ↗official | |
| DystopiaBenchlaguardia_score | #31 / 50 | ↓68.27 | Source ↗official | |
| DystopiaBenchorwell_score | #35 / 50 | ↓73.53 | Source ↗official | |
| DystopiaBenchpetrov_score | #28 / 50 | ↓73.53 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #51 / 241 | ↑27.65 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #194 / 241 | ↑81 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #82 / 241 | ↑80.56 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #220 / 239 | ↑80 | Source ↗official | |
| FORTRESSaverage_risk_score | #15 / 49 | ↓17.62 | Source ↗official | |
| FORTRESSover_refusal_score | #43 / 48 | ↓11.94 | Source ↗official | |
| HELM Safetyanthropic_red_team | #10 / 80 | ↑0.997 | Source ↗official | |
| HELM Safetybbq | #13 / 80 | ↑0.967 | Source ↗official | |
| HELM Safetyharmbench | #2 / 80 | ↑0.987 | Source ↗official | |
| HELM Safetysimple_safety_tests | #1 / 80 | ↑1 | Source ↗official | |
| HELM Safetyxstest | #72 / 80 | ↑0.872 | Source ↗official | |
| Manager Coercion Benchcoercion_ladder_depth | #11 / 33 | ↓7.9 | Source ↗self run | |
| MASKlying_probability_pct | #10 / 53 | ↓13.54 | Source ↗official | |
| ODCV-Benchaverage_severity | #7 / 12 | ↓1.475 | Source ↗official | |
| ODCV-Benchmisalignment_rate | #5 / 12 | ↓28.75 | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #65 / 113 | ↑6.75 | Source ↗official | |
| SpeciEvalland_animal_4ns | #15 / 113 | ↓4.2 | Source ↗official | |
| SpeciEvalsea_animal_4ns | #68 / 113 | ↓4.83 | Source ↗official | |
| SpeciEvalspeciesism | #45 / 113 | ↓1.9 | Source ↗official | |
| Vigil Mental Health Safetyoverall_score | #17 / 23 | ↑31 | Source ↗official |
Values evaluations
Descriptive values and political-framing results are separate from safety/ethics ranks. Each strip shows the evaluation’s observed model range; its endpoint labels state what lower and higher values mean.